Multiple Alterations in Insulin Responses to Glucose in Islets From 48-h Glucose-Infused Nondiabetic Rats

To examine the biochemical mechanisms by which hyperglycemia produces insulin secretory abnormalities, we studied isolated islets from control rats and rats infused for 48 h with a 50% glucose solution. To preserve the effects of in vivo hyperglycemia during in vitro handling for islet isolation, our standard isolation procedure utilized buffers containing 16.8 mM glucose. Islets from infused rats released similar amounts of insulin in low or high glucose during first incubations at 37°C (92.4 ± 7.0 ng 10 islets1 45 min1 at 2.8 mM, 84.4 ± 4.1 ng · 10 islets⁻¹ · 45 min⁻¹ at 16.8 mM) in contrast with control (uninfused) islets (18.6 ± 2.8 ng 10 islets1 · 45 min1 at 2.8 mM and 109.8 ± 8.0 ng 10 islets1 45 min1 at 16.8 mM glucose) (P < 0.01). Secretion by islets of glucose-infused rats was lower during 60-min second incubations at 28 mM glucose than in first incubations of the same islets in low glucose (P < 0.01). This phenomenon is comparable to the paradoxical hypersecretion observed during the first 10–15 min of exposure of glucose-infused pancreas to low-glucose perfusions. Paradoxical secretion in low glucose waned rapidly, so that during second incubations at 37°C, little immunoreactive insulin release occurred at 2.8 mM glucose, despite the persistence of two additional lesions. The glucoseinsulin dose-response curves in second incubations showed a leftward shift for glucose-infused islets, with two-to threefold higher secretion at 5.6–8.4 mM glucose than control islets. This is termed sensitization to glucose. Furthermore, infused islets showed an impaired insulin release (42–73% of control islets) at high glucose (16–28 mM); the decreased maximum response to glucose is related to the previously described desensitization to glucose. This abnormality was reversed by isolation of islets in glucose-free buffer, whereas paradoxical hypersecretion in first incubations was partially reversed, and sensitization to glucose remained undiminished. Studies of secretory responses to nonglucose secretagogues indicated that the maximum response to D-glyceraldehyde was impaired in infused islets, whereas the responses to α-ketoisocaproate and tolbutamide were not significantly different from control. Both monooleoylglycerol, an inhibitor of the turnover of intracellular diacylglycerol, and 12-O-tetradecanoylphorbol-13-acetate were able to restore the decreased maximal secretion in high glucose to control levels. However, monooleoylglycerol had little effect on the low response to glyceraldehyde. Insulin secretion by infused islets was resistant to both mannoheptulose (0.75–3.0 mM) and verapamil (5 μM) inhibition in the presence of high glucose compared with control islets. The data on reversal support the existence of three separate lesions in insulin secretory responses to glucose and suggest that multiple intracellular regulatory mechanisms have been altered. Decreased maximal response to high glucose (desensitization) is the lesion that is most clearly related to and maintained by hyperglycemia, whereas the other two lesions may be related to hypersecretion per se or to extraislet effects in vivo.